Skip to content

amokeev/ru_norm_kaggle

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ru_norm_kaggle: Text Normalization Challenge - Russian Language

This is the winning solution to Text Normalization Challenge - Russian Language

The project is composed of two subprojects:

  • train
  • translation

The first is used to train the model. Its output is a binary model file. The second is used to apply trained model to the input and do postprocessing.

Please checkout out corresponding READMEs in these projects for details on theirs execution.

REFERENCE FILES

These files were generated by the projects above. They can be used to jump over a few steps.

The files are in this folder on Goggle Drive

  • model.t7 - trained model, you can use it to translate
  • dictionary.csv - full dictionary, created according to the steps from translation/README
  • restored.csv - translation output, w/o any postprocessing. You can use it to verify your training+translation procedure
  • restored.pp.csv - postprocessed translation output, equal to the winning submission

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published